I had a look at the documentation and publication reference. It seems that PsPM is specialized for event-related designs. However, we have a block design (17 s) with 4 conditions. Which of the First Level models available in PsPM would be the most suitable for this case? Would you suggest to use the "GLM SCR" model and define the 17-s duration for the blocks? Or would it be more suitable to use the SF (spontaneous fluctuation) model?
We also have ECG data collected. Does it make sense to use the "GLM for HP" module for the 17-s block design or would this stress unduly the model?
Thanks in advance.
Sebastian
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
sorry for the delay in responding - summer holiday time.
Long-lasting stimuli don't normally elicit continuous sudomotor firing (unlike what you may expect in an fMRI model), therefore a "boxcar" or "block" regressor would not capture a physiological reality.
The most likely situation is that your stimuli elicit sudomotor bursts. These could be analysed with the SF models, and on top of analysing the frequency of bursts (which PsPM outputs automatically) , you could even look at their amplitude and timing.
For ECG, I think there may be tonic autonomic inputs but their impact is likely to differ from phasic input - so I would be sceptical about an ECG model with "boxcar" regressors.
Hope this helps
Dominik
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear Dr. Bach,
I had a look at the documentation and publication reference. It seems that PsPM is specialized for event-related designs. However, we have a block design (17 s) with 4 conditions. Which of the First Level models available in PsPM would be the most suitable for this case? Would you suggest to use the "GLM SCR" model and define the 17-s duration for the blocks? Or would it be more suitable to use the SF (spontaneous fluctuation) model?
We also have ECG data collected. Does it make sense to use the "GLM for HP" module for the 17-s block design or would this stress unduly the model?
Thanks in advance.
Sebastian
Dear Sebastian
sorry for the delay in responding - summer holiday time.
Long-lasting stimuli don't normally elicit continuous sudomotor firing (unlike what you may expect in an fMRI model), therefore a "boxcar" or "block" regressor would not capture a physiological reality.
The most likely situation is that your stimuli elicit sudomotor bursts. These could be analysed with the SF models, and on top of analysing the frequency of bursts (which PsPM outputs automatically) , you could even look at their amplitude and timing.
For ECG, I think there may be tonic autonomic inputs but their impact is likely to differ from phasic input - so I would be sceptical about an ECG model with "boxcar" regressors.
Hope this helps
Dominik